Skip to content

Commit

Permalink
completing narrative
Browse files Browse the repository at this point in the history
  • Loading branch information
@k8hertweck
k8hertweck committed Apr 8, 2015
1 parent e1ce621 commit 70d645e
Showing 1 changed file with 16 additions and 26 deletions.
42 changes: 16 additions & 26 deletions labExercises/BioinfWk12Lab.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -19,45 +19,33 @@ Additional materials for reference:

**Load and filter data**

Navigate to the [Galaxy website](https://usegalaxy.org). Log in and click on the button that looks like an open book in the top right hand corner of the screen to view your histories. Create a new history called "metagenomics" and make sure it's your current history before navigating back to your work page (click "Analyze Data" at the top of the screen).
*Create new history* Navigate to the [Galaxy website](https://usegalaxy.org). Log in and click on the button that looks like an open book in the top right hand corner of the screen to view your histories. Create a new history called "metagenomics" and make sure it's your current history before navigating back to your work page (click "Analyze Data" at the top of the screen).

Select "Data Libraries" in the "Shared Data" drop down menu at the top of the screen. Search for "windshield" and click on the result to view the data files. Click on the boxes next to "Trip A left Side Reads" and "Trip A Left Side QV" and then hit Go to import these data to your current history.
*Load data* Select "Data Libraries" in the "Shared Data" drop down menu at the top of the screen. Search for "windshield" and click on the result to view the data files. Click on the boxes next to "Trip A left Side Reads" and "Trip A Left Side QV" and then hit Go to import these data to your current history.

If you look back at your work space, the "reads" is the file containing the fasta sequences, while "QV" is a separate file containing quality scores.

Under "NGS: QC and manipulation," choose "Select high quality segments." Galaxy should autodetect and enter "454 reads" in the "Reads" field and "454 qualities" in the "Quality scores" field for you. Select a minimal quality score of 20 and minimal length of contiguous segment as 50. 454 should be pre-selected for you under "Select technology," and leave the default option for "DO NO trigger splitting." While this doesn't report exactly how many sequences were filtered out, the file size does decrease.
*Quality filter reads* Under "NGS: QC and manipulation," choose "Select high quality segments." Galaxy should autodetect and enter "454 reads" in the "Reads" field and "454 qualities" in the "Quality scores" field for you. Select a minimal quality score of 20 and minimal length of contiguous segment as 50. 454 should be pre-selected for you under "Select technology," and leave the default option for "DO NO trigger splitting." While this doesn't report exactly how many sequences were filtered out, the file size does decrease.

"FASTA to Tabular" under "Convert formats" to convert to a tab-delimited file format, leave number of columns as 1?

Under "Text Manipulation," select "Add column". "TripA" as value and "YES" to iterate, should autodetect dataset.

Under "Convert Formats" select "Tabular-to-FASTA", tab-delimited file should auto-detect as the column added file, title column is 3 and sequence column is 2.
*Add sample label* Now we are going to perform some file manipulations to add the sample label, indicating Trip A, to our data file. Go to "FASTA to Tabular" under "Convert formats" to convert to a tab-delimited file format, leaving the number of columns as 1. Then, under "Text Manipulation," select "Add column". Enter "TripA" as the value and "YES" to iterate (the filtered dataset should be autodetected). Now that this information is added, we need to convert back to FASTA. Under "Convert Formats" select "Tabular-to-FASTA". The tab-delimited file should auto-detect as the column added file, and select title column as 3 and sequence column as 2.

**Sequence searching**

Under "NGS: Mapping" select "Megablast." Should autodetect dataset as output from tabular-to-FASTA. Select "wgs" for "against target database," 28 as word size, 80 as report hit threshold, and 0.0001 as e-value cutoff. What do the columns in the output file mean?

Perform another "Megablast" with the same parameters, except select "nt" as the target database.
*Megablast* Under "NGS: Mapping" select "Megablast." Should autodetect dataset as output from tabular-to-FASTA. Select "wgs" for "against target database," 28 as word size, 80 as report hit threshold, and 0.0001 as e-value cutoff. Perform another "Megablast" with the same parameters, except select "nt" as the target database.

Under "FASTA manipulation," select "Compute sequence length" on the tabular-to-fasta converted data.
*Manually upload Megablast results, if jobs take too long* Running megablast is a big job, so we're going to take a shortcut and upload example results files (which should be identical to the results your jobs will get after they run). Download megablast.zip from the [GitHub BioinformaticsSpring2015/data folder](https://github.com/BioinformaticsSpring2015/BioinformaticsMaterials/tree/master/data). Unzip the file. Go back to Galaxy and select "Upload File" under the "Get Data" section in the lefthand toolbar. Drag the files to the window and click "Start" to upload these files. The files will appear in the righthand toolbar. When they turn green (meaning loaded and available), preview the data in one of the files. What data is included in each column?

Under "Text Manipulation," select "Concatenate datasets" to combine the results from the two megablast searches.

Under "Join, Subtract, and Group," select "Join two Datasets" to combine your concatenated megablast results with the computed sequence lengths (use default values)

Under "Filter and sort," select "Filter." The megablast with sequence lengths should be autodetected as the file to filter. Use the following condition to filter: "c5/c15 < 0.5"
*Concatenate megablast searches and filter by sequence length* Under "Text Manipulation," select "Concatenate datasets" to combine the results from the two megablast searches. Next, create a new tabular dataset including the length of each sequence by going to "FASTA manipulation" and running "Compute sequence length" on the tabular-to-fasta converted data. Add the sequence length to the megablast results by going to the "Join, Subtract, and Group" category and selecting "Join two Datasets." Join your concatenated megablast results with the computed sequence lengths on column 1 (unique names for each sequence). Finally, under "Filter and sort," select "Filter." The megablast results with sequence lengths should be autodetected as the file to filter. Use the following condition to filter: "c5/c15 < 0.5". This filter removes megablast hits that cover less than 50% of the sequence read. What data columns are included in this filter?

**Metagenomic analysis**

The rest of our analyses will use commands under the "Metagenomics analysis" section of the tools sidebar.

Select "Fetch taxonomic representation." Input your filtered data file. Select 2 for the GIs column, and 1 for Name column.

Select "Find lowest diagnostic rank" and input your taxonomic data file. Require the lowest rank to be at least Subkingdom.
First, select "Fetch taxonomic representation." Input your filtered data file (should autodetect). Select 2 for the GIs column, and 1 for Name column. What does this command do?

Select "Summarize taxonomy" and to summarize your taxonomic representation file.
Next, select "Find lowest diagnostic rank" and input your taxonomic data file (from last step, should autodetect). Require the lowest rank to be at least Subkingdom. This command reconciles differing blast results to the most specific identification possible.

Select "Summarize phylogeny" on the summarized taxonomy file, showing ranks from root to Class.
Then select "Summarize taxonomy" to count the sequences from each taxnomic group in your results. Finally, select "Summarize phylogeny" on the summarized taxonomy file, showing ranks from root to Class. How do you interpret these results?

###Assignment
* Due Wednesday, April 15 at 5 pm
Expand All @@ -74,10 +62,12 @@ Select "Summarize phylogeny" on the summarized taxonomy file, showing ranks from
* Don't forget to preview your homework before committing!
* If you get stuck on a question, please consult readings above.

1. What does Megablast do?
2. Change parameters and run
3. Why did we add sequence length?
4. Why did we add "TripA" to the sequence IDs?
1. What does megablast do? Why did we use it for this lab?
2. Why did we add "TripA" to the sequence IDs?
3. Why did we calculate sequence length and add it to the megablast results?
4. How would you expect your final results to differ if you do not filter the data? Now re-run the last few steps on the unfiltered dataset. Was your hypothesis supported?
3.
4.


8. How long did it take you to complete these questions?
Expand Down

0 comments on commit 70d645e

Please sign in to comment.